CN115168554A - Callback object return visit method and device, storage medium and computer equipment - Google Patents
Callback object return visit method and device, storage medium and computer equipment Download PDFInfo
- Publication number
- CN115168554A CN115168554A CN202210736358.5A CN202210736358A CN115168554A CN 115168554 A CN115168554 A CN 115168554A CN 202210736358 A CN202210736358 A CN 202210736358A CN 115168554 A CN115168554 A CN 115168554A
- Authority
- CN
- China
- Prior art keywords
- target
- outbound object
- outbound
- preset
- call
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/338—Presentation of query results
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/63—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for estimating an emotional state
Abstract
The invention discloses a callback method, a callback device, a storage medium and computer equipment, relates to the technical field of communication, and mainly aims to enable the callback method to be suitable for a callback object so as to improve the callback accuracy of the callback object. The method comprises the following steps: acquiring characteristic data of an outbound object; determining a first target call group to which the outbound object belongs based on the feature data; based on a first preset dialect corresponding to the first target calling group, carrying out intelligent communication with the outbound object to obtain a returned voice fed back by the outbound object aiming at the first preset dialect; determining a target return visit method corresponding to the outbound object based on the returned voice; and based on the target return visit method, returning visits to the outbound object. The invention is suitable for revisiting the outbound object.
Description
Technical Field
The present invention relates to the field of communications technologies, and in particular, to a callback object return access method, apparatus, storage medium, and computer device.
Background
In the big data era, the outbound system is endlessly developed, the main function is to make a machine replace manpower to communicate with a customer, the outbound system is an indispensable important component in many complex product services, and is an important channel for enterprise activity popularization, for example: insurance business services, financial services, real estate services, automotive services, curriculum services, travel route business services, activity invitations, and the like. In each business scenario, an outbound call may not facilitate the order, and therefore, it becomes important to return the customer to the call in order to increase the volume of the order.
Currently, a return visit to a client is typically accomplished by making the same outgoing call to the client multiple times. However, this method causes the uninteresting client to receive the same call for many times, which may result in the client being disgust, and the same outbound call for many times may cause the uninteresting client not to receive the desired return visit, which may result in a lower accuracy of the return visit to the outbound call subject.
Disclosure of Invention
The invention provides a callback method, a callback device, a storage medium and computer equipment, and mainly aims to enable the callback method to be suitable for a callback object so as to improve the callback accuracy of the callback object.
According to a first aspect of the present invention, there is provided a callback method for an outbound subject, comprising:
acquiring characteristic data of an outbound object;
determining a first target call group to which the outbound object belongs based on the feature data;
based on a first preset dialect corresponding to the first target calling group, carrying out intelligent communication with the outbound object to obtain a returned voice fed back by the outbound object aiming at the first preset dialect;
determining a target return visit method corresponding to the outbound object based on the returned voice;
and based on the target return visit method, returning visits to the outbound object.
According to a second aspect of the present invention, there is provided an apparatus for revisiting an outbound subject, comprising:
the acquiring unit is used for acquiring the characteristic data of the outbound object;
a first determining unit, configured to determine, based on the feature data, a first target call group to which the outbound object belongs;
the intelligent communication unit is used for carrying out intelligent communication with the outbound object based on a first preset dialect corresponding to the first target calling group to obtain a returned voice fed back by the outbound object aiming at the first preset dialect;
a second determining unit, configured to determine, based on the returned speech, a target return visit method corresponding to the outbound object;
and the return visit unit is used for returning visits to the outbound object based on the target return visit method.
According to a third aspect of the present invention, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
acquiring characteristic data of an outbound object;
determining a first target call group to which the outbound object belongs based on the feature data;
based on a first preset dialect corresponding to the first target calling group, carrying out intelligent communication with the outbound object to obtain a returned voice fed back by the outbound object aiming at the first preset dialect;
determining a target return visit method corresponding to the outbound object based on the returned voice;
and based on the target return visit method, returning visits to the outbound object.
According to a fourth aspect of the present invention, there is provided a computer apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the program:
acquiring characteristic data of an outbound object;
determining a first target call group to which the outbound object belongs based on the feature data;
based on a first preset dialect corresponding to the first target calling group, carrying out intelligent communication with the outbound object to obtain a returned voice fed back by the outbound object aiming at the first preset dialect;
determining a target return visit method corresponding to the outbound object based on the returned voice;
and based on the target return visit method, returning visits to the outbound object.
According to the callback object return visit method, the callback object return visit device, the storage medium and the computer equipment, compared with the current mode of returning the client by carrying out the same callback for a plurality of times, the callback object return visit method, the device and the storage medium have the advantages that the characteristic data of the callback object are obtained; determining a first target calling group to which the outbound object belongs based on the characteristic data; meanwhile, based on a first preset dialect corresponding to the first target calling group, carrying out intelligent communication with the outbound object to obtain a returned voice fed back by the outbound object aiming at the first preset dialect; then, based on the returned voice, determining a target return visit method corresponding to the outbound object; finally, based on the target return visit method, the outbound object is returned, so that a first target calling group corresponding to the outbound object is determined, communication is carried out with the outbound object based on a first preset conversation corresponding to the first target calling group, then a target return visit method corresponding to the outbound object is determined based on returned voice obtained through the communication, and finally, the outbound object is returned based on the target return visit method, so that the situation that a user is called for many times to cause boredom of the user can be avoided, meanwhile, the time consumed by calling the user for many times can be reduced, meanwhile, the call with the same content can be avoided from being called for many times, the client who wants to handle the service does not obtain effective return visit content, and the return visit accuracy of the outbound object is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
fig. 1 shows a flowchart of a callback method of an outbound object according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating another callback method for an outbound object according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram illustrating an apparatus for revisiting an outbound object according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of another callback device for an outbound object according to an embodiment of the present invention;
fig. 5 shows a schematic physical structure diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The invention will be described in detail hereinafter with reference to the drawings and embodiments. It should be noted that, in the present application, the embodiments and features of the embodiments may be combined with each other without conflict.
At present, the mode of calling the clients back by calling the clients for a plurality of times is the same, so that the clients who are not interested receive the same call for a plurality of times, the clients who are interested do not obtain the wanted call back content, and the call back effect is poor.
In order to solve the above problem, an embodiment of the present invention provides a callback method for an outbound object, as shown in fig. 1, where the method includes:
101. and acquiring characteristic data of the outbound object.
The outbound object is a called party in the call service system, for example, the outbound object may be a user who purchases a financial product, a user who selects an automobile, a user who purchases a car insurance, or the like, the outbound object may be one outbound object or a plurality of outbound objects, and the feature data of the outbound object includes age, gender, occupation, income data, interests and hobbies, and the like.
For the embodiment of the invention, in order to solve the problem of low return visit accuracy of the outbound object in the prior art, the embodiment of the invention determines the first target call group corresponding to the outbound object, communicates with the outbound object based on the first preset dialect corresponding to the first target call group, then determines the target return visit method corresponding to the outbound object based on the return voice obtained by the communication, and finally returns visit to the outbound object based on the target return visit method, so that the situation that the user is bored due to calling for many times can be avoided, the time consumed for calling for the user for many times can be reduced, meanwhile, the situation that the user who wants to handle the service does not obtain effective return visit content can be avoided, and the return visit accuracy of the outbound object can be improved. The embodiment of the invention is mainly applied to a scene of revisiting an outbound object, and the execution main body of the embodiment of the invention is a device or equipment capable of revising the outbound object, and can be specifically arranged on one side of a client or a server.
Specifically, a company where a calling party is located preliminarily studies and collects characteristic data of age, sex, occupation, income and the like of each outbound object, the collected characteristic data and an Identity (ID) of the outbound object are used as identifiers and are stored in a database of the company, if the characteristic data of the outbound object are required to be obtained, the ID of the outbound object is used as a search condition, the characteristic data of the outbound object can be searched in the database, a first target calling group of the outbound object is determined based on the characteristic data of the outbound object, the first target calling group corresponding to the outbound object is determined, and a conversation is carried out with the outbound object by using a conversation technique corresponding to the first target calling group, so that the requirements of customers can be met, and the order volume is increased.
102. And determining a first target call group to which the outbound object belongs based on the characteristic data.
The different service scenes correspond to different call groups, for example, the insurance service scene corresponds to an insurance call group, the medical service scene corresponds to a medical call group, and the car purchasing service scene corresponds to a vehicle call group.
For the embodiment of the present invention, after the feature data of the calling object is obtained, the feature data of the calling object is analyzed, and the first target calling group to which the feature object belongs is determined, for example, if the feature data of the calling object includes age 25, profession of engineers, income of 20 ten thousands yuan, and hobbies and interests of tourism, based on the feature data, it can be analyzed that the first target calling group of the calling object is a vehicle service group, and intelligent communication is performed with the outbound object based on the first preset word technique corresponding to the vehicle service group, so that customers of different calling groups communicate with the calling object by using different preset words techniques, the requirements of the customers can be better satisfied, the dislike of the customers is reduced, and the volume of orders is increased.
103. And intelligently communicating with the outbound object based on a first preset dialect corresponding to the first target calling group to obtain a returned voice fed back by the outbound object aiming at the first preset dialect.
The first preset dialect may be a idiom for making an outbound from an AI intelligent outbound platform, different calling groups correspond to different idioms, specific contents of the idioms may be set based on a service scene, the embodiment of the present invention is not particularly limited, and the returned voice is a speech fed back by the outbound object in a call process.
For the embodiment of the present invention, after determining the first target call group corresponding to the outbound object, in order to return the outbound object, first a first preset telephone operation corresponding to the first target call group needs to be determined, and a telephone number and a name corresponding to the outbound object are determined at the same time, through the telephone number, the AI intelligence outbound system automatically connects to the outbound object according to the telephone number, when the outbound object connects to the connection, the AI intelligence outbound system calls to the outbound object according to the first preset telephone operation according to the name of the outbound object, and at the same time, the AI intelligence outbound system automatically records a recording of the call with the outbound object, after the call is ended, selects a voice of the outbound object from the recording at this time, that is a voice returned by the outbound object to the first preset telephone operation, for example, if the first target call group to which the outbound object belongs to is an insurance service, it is determined that the first preset telephone operation corresponding to the insurance service is a "hello call group", and we have a recent requirement for buying a new insurance company XX, and we have a new insurance company. When a calling object answers a call, an outbound robot in the AI intelligent outbound system intelligently communicates with the outbound object according to the first preset dialect, records returned voice fed back by the outbound object aiming at the first preset dialect, then determines a target return visit method corresponding to the outbound object based on the returned voice, and finally returns a visit to the outbound object based on the target return visit method, so that the situation that the client is bored due to the fact that the user calls for many times can be avoided, and meanwhile, the time consumed by the user to call for many times can be reduced.
104. And determining a target return visit method corresponding to the outbound object based on the returned voice.
For the embodiment of the invention, after the returned voice of the outbound object is obtained, the returned voice is analyzed, the emotion of the outbound object can be judged through the returned voice, if the outbound object obviously does not like to answer the call, the outbound object can be added into a blacklist, that is, the outbound object does not need to be revisited later, if the emotion of the outbound object answering the call is good, the requirement of the outbound object can be continuously judged through the returned voice, and the target revising method needed by the outbound object is determined based on the requirement of the outbound object, and finally the outbound object is revisited based on the target revising method, so that the condition that the same content of the call is dialed for the client for multiple times can be avoided, the client wanting to handle the service does not obtain effective revising content, and the revising accuracy of the outbound object is improved.
105. And based on the target return visit method, returning visits to the outbound object.
For the embodiment of the invention, after the target return visit method of the outbound object is obtained, the outbound object is returned according to the target return visit method, for example, if the target return visit method of the outbound object is determined to be a transfer artificial customer service through return voice of the outbound object, the AI intelligent outbound system directly switches the outbound call conversation process to an artificial platform under the service scene, and the artificial platform is manually butted to perform next more detailed communication, so as to meet the requirements of the client and improve the call volume.
According to the method for revisiting the outbound object provided by the invention, compared with the mode of revising the client by carrying out the same outbound for a plurality of times on the client at present, the method for revisiting the outbound object obtains the characteristic data of the outbound object; determining a first target calling group to which the outbound object belongs based on the characteristic data; meanwhile, based on a first preset dialect corresponding to the first target calling group, carrying out intelligent communication with the outbound object to obtain a returned voice fed back by the outbound object aiming at the first preset dialect; then, based on the returned voice, determining a target return visit method corresponding to the outbound object; finally, based on the target return visit method, the outbound object is returned, so that a first target calling group corresponding to the outbound object is determined, communication is carried out with the outbound object based on a first preset conversation corresponding to the first target calling group, then a target return visit method corresponding to the outbound object is determined based on returned voice obtained through the communication, and finally, the outbound object is returned based on the target return visit method, so that the situation that a user is called for many times to cause boredom of the user can be avoided, meanwhile, the time consumed by calling the user for many times can be reduced, meanwhile, the call with the same content can be avoided from being called for many times, the client who wants to handle the service does not obtain effective return visit content, and the return visit accuracy of the outbound object is improved.
Further, to better explain the procedure of revisiting the outbound object, as a refinement and an extension to the foregoing embodiment, an embodiment of the present invention provides another revisiting method for the outbound object, as shown in fig. 2, where the method includes:
201. characteristic data of the outbound object is obtained.
For the embodiment of the invention, the characteristic data of the age, sex, occupation, income, interest and hobby of the outbound object is obtained in a company database through the identity ID of the outbound object, the first target calling group to which the outbound object belongs is determined based on the characteristic data, then the first preset telephone corresponding to the first target calling group is determined, and simultaneously the information of the telephone number and the like corresponding to the outbound object is uploaded to an AI intelligent outbound system, and the AI intelligent outbound system is in intelligent communication with the outbound object based on the telephone number and the first preset telephone.
202. And determining a first target calling group to which the outbound object belongs based on the characteristic data.
For the embodiment of the present invention, in order to determine the first target call group to which the outbound object belongs, different call groups need to be created according to different service scenarios, and based on this, the method includes: establishing corresponding call groups for different service scenes; and creating preset dialogs corresponding to the different call groups.
Specifically, different call groups are created according to business scenarios, the different business scenarios have different call groups, for example, an insurance call group is created for an insurance business scenario, a medical call group is created for a medical business scenario, a real estate call group is created for a real estate business scenario, a travel call group is created for a travel business scenario, and meanwhile, in order to meet the needs of customers and increase the user's interest, it is further required to create preset dialogues corresponding to the different call groups, for example, the preset dialogues related to insurance are created for the insurance call group, the preset dialogues related to medical treatment are created for the medical call group, the preset dialogues related to real estate are created for the real estate call group, and meanwhile, the different business scenarios all have corresponding customer characteristics, for example, customers corresponding to the medical business scenarios are generally people aged over 50 years, customers corresponding to real estate business scenarios are generally people aged over 20 years, customers corresponding to vehicle business scenarios are generally aged over 25-40 years, the earnings corresponding to the foreign countries are called based on the first calling characteristic data, and the first calling object is determined based on the first step: and inputting the characteristic data into a preset group division model for group division to obtain a first target calling group to which the outbound object belongs.
Specifically, after obtaining feature data corresponding to the outbound object, in order to determine a first target call group corresponding to the outbound object, the feature data needs to be input into a preset group division model for group division, in an optional embodiment of the present invention, before inputting the feature data into the preset group division model for group division, the preset group division model may be established and trained first, and the method for specifically training and establishing the preset group division model is that historical order data of a call object that has completed a call is collected first, which refers to a call record of a user who has completed a call, the call record includes sample feature data corresponding to the call object that has completed a call, and a sample call group to which the call object that has completed a call belongs, and the sample feature data and the sample call group corresponding thereto are used as a training set, training and constructing the preset group division model, wherein the training process comprises the steps of inputting the sample characteristic data into the preset initial group division model for group division to obtain a predicted calling group corresponding to the calling object which finishes calling, comparing the predicted calling group with the sample calling group to construct a loss function corresponding to the preset initial group division model, and finally training and constructing the preset group division model based on the loss function, wherein the preset group division model can be specifically a preset multilayer perceptron, the preset multilayer perceptron comprises an input layer, a hidden layer and an output layer, inputting the characteristic data of the calling object into the hidden layer through the input layer of the preset multilayer perceptron, and outputting a first characteristic vector corresponding to the calling object through the hidden layer, then, the first feature vector is input to the output layer, the output layer outputs a first target call group to which the call object belongs, and then a first preset session corresponding to the first target group is determined, for example, if the first target call group of the call object obtained by using the preset group division model is a medical call group, the first preset session corresponding to the medical call group can be determined as the medical preset session based on a corresponding relationship between a call group owner and the preset session.
203. And intelligently communicating with the outbound object based on a first preset speech technology corresponding to the first target call group to obtain returned voice fed back by the outbound object aiming at the first preset speech technology.
For the embodiment of the present invention, in order to utilize the first preset conversation to communicate with the outbound object, it is first required to determine whether the outbound object receives the conversation, and based on this, the method includes: responding to a call result of calling the outbound object, and judging whether the call result is an answering call; if the calling result is answering the call, intelligently communicating with the outbound object based on a first preset dialect corresponding to the first target calling group to obtain returned voice fed back by the outbound object aiming at the first preset dialect; and if the calling result is that the call is refused to be answered, determining a calling period corresponding to the outbound object, and calling the outbound object again based on the calling period.
For the embodiment of the present invention, after determining a first target call group to which the outbound object belongs and a first preset call corresponding to the first target call group, establishing a communication connection with the outbound object based on a communication number of the outbound object, that is, dialing a phone number of the outbound object, calling the outbound object, monitoring whether the call object answers the call, if the call object does not answer the call, setting a call cycle of the outbound object, and calling the outbound object again based on the call cycle, for example, if the outbound object does not answer the call, setting the call cycle to be 5 days and 3 times, that is, making 3 calls for a user who answers the call for no 5 days until the user makes a call to a phone position, if the outbound object hangs up the phone during the process of making a call to the outbound object, determining that the outbound object does not want to receive such a phone, then the outbound object needs to be listed in a blacklist, and in order to not cause trouble to the client to make a call, the client is not bothered.
Further, after the outbound object agrees with the communication connection, the outbound object communicates with the outbound object based on a first preset dialect corresponding to the first target group, in the process, recording the whole process recording of the communication with the outbound object, screening out the voice of the outbound object in the recording, obtaining the returned voice fed back by the outbound object aiming at the first preset dialect, and finally determining a target return visit method corresponding to the outbound object based on the returned voice.
204. And performing word segmentation processing on the voice text corresponding to the returned voice to obtain each word segmentation corresponding to the voice text.
Specifically, after the return speech fed back by the outbound object for the first preset dialect is obtained, in order to determine a target return visit method corresponding to the outbound object, word segmentation processing needs to be performed on a speech text corresponding to the return speech to obtain each word segmentation corresponding to the return speech, for example, if the speech text corresponding to the return speech is: "i need to communicate with the manual customer service further", performing word segmentation processing on the voice text to obtain each word segmentation corresponding to the voice text: "i/need/and/manual/customer service/go/communicate/".
205. And removing stop words in the participles to obtain the key participles with the stop words removed.
Specifically, after obtaining each participle corresponding to the voice text, in order to reduce the computational complexity, it is necessary to detect whether each participle includes a stop word in the preset stop word library, and if each participle includes a stop word in the preset stop word library, the stop word in each analysis is removed to obtain a key participle corresponding to the returned voice.
206. And determining a target preset word bank to which the key word segmentation belongs.
Specifically, a preset lexicon corresponding to different user requirements is established in advance, for example, the preset lexicon is a manual customer service preset lexicon, the preset lexicon is a rejection requirement preset lexicon, the preset lexicon is a service scene preset lexicon, the artificial customer service preset lexicon comprises words such as transfer workers, customer service and the like, the rejection requirement preset lexicon comprises words such as unnecessary words, bad words and the like, the service scene preset lexicon comprises words such as medical words, vehicles, insurance and the like, meanwhile, different preset lexicons correspond to different return visit methods, for example, the return visit method corresponding to the manual customer service preset lexicon is to transfer a call between a manual customer and the call, the return visit method corresponding to the rejection requirement preset lexicon is to not return the call to the customer, the return visit method corresponding to the service scene preset lexicon is to re-determine a target call group to which the call object belongs, and to call the call object again based on the preset lexicon corresponding to the target call object, for the embodiment of the invention, after determining the key participles corresponding to the return speech, the target lexicon belongs to the target call object is determined, and the return visit method corresponding to the target object is finally based on the return visit method.
207. And determining a return visit method corresponding to the target preset word bank as a target return visit method corresponding to the outbound object.
Specifically, after determining the return visit method corresponding to the target preset lexicon, the return visit method corresponding to the target preset lexicon needs to be determined as the target return visit method corresponding to the call object, and finally the call object is returned based on the target return visit method, for example, if a key word segmentation corresponding to the return voice is an operation, the target preset lexicon to which the key word segmentation belongs is determined as a service scene preset lexicon, meanwhile, based on the operation key word, the call object can be determined as a medical service scene, the call object can be determined as a medical call group, a power ring needs to be hung up, and then based on a preset word corresponding to the medical call group, a telephone of the call object is dialed up again to communicate with the call object.
208. And based on the target return visit method, the return visit is carried out on the outbound object.
For the embodiment of the present invention, after determining the target return visit method of the outbound object, the outbound object needs to be returned visit based on the target return visit method, and based on this, step 208 specifically includes: if the target preset lexicon is a first preset lexicon, switching to the artificial customer service corresponding to the first target call group to communicate with the outbound object; and if the target preset word bank is a second preset word bank, storing the outbound object to a blacklist, and stopping the return visit of the outbound object.
Specifically, the first preset lexicon is a preset lexicon for artificial customer service, the second preset lexicon is a preset lexicon for rejecting demand, if the target preset lexicon is the preset lexicon for artificial customer service, it can be determined that the outbound object wants to communicate with the artificial customer service, at this time, communication is automatically switched to the artificial customer service corresponding to the target call group, the artificial customer service is allowed to communicate with the call object, if the target preset lexicon is the preset lexicon for rejecting demand, it can be determined that the outbound object does not want to answer such sales calls, return visits are performed to the outbound object, the dislike of the outbound object will be increased, at this time, the outbound object needs to be added into a blacklist, that is, the return call is not performed to the outbound object.
Further, if the target preset lexicon is a third preset lexicon, the callback object needs to be revisited based on a revisiting method corresponding to the third preset lexicon, and based on this, step 208 specifically includes: if the target preset word bank is a third preset word bank, re-determining a second target call group to which the outbound object belongs based on the key participle; and carrying out intelligent communication with the outbound object based on a second preset conversation corresponding to the second target calling group.
Specifically, the third preset lexicon is a preset lexicon of a service scene, if it is determined that the returned speech of the outbound object contains words of other service scenes, such as a medical service scene, the outbound object is re-divided into a medical call group, and the outbound object is called again based on a communication number of the call object, and if the outbound object answers the call, the third preset lexicon is in intelligent communication with the outbound object based on a second preset terminology corresponding to the medical call group.
According to another callback object return visit method provided by the invention, compared with the current mode of carrying out the same callback for a plurality of times on the client to realize the return visit to the client, the method provided by the invention obtains the characteristic data of the callback object; determining a first target calling group to which the outbound object belongs based on the characteristic data; meanwhile, based on a first preset dialect corresponding to the first target calling group, carrying out intelligent communication with the outbound object to obtain a returned voice fed back by the outbound object aiming at the first preset dialect; then, based on the returned voice, determining a target return visit method corresponding to the outbound object; finally, based on the target return visit method, the outbound object is returned, so that a first target calling group corresponding to the outbound object is determined, communication is carried out with the outbound object based on a first preset conversation corresponding to the first target calling group, then a target return visit method corresponding to the outbound object is determined based on returned voice obtained through the communication, and finally, the outbound object is returned based on the target return visit method, so that the situation that a user is called for many times to cause boredom of the user can be avoided, meanwhile, the time consumed by calling the user for many times can be reduced, meanwhile, the call with the same content can be avoided from being called for many times, the client who wants to handle the service does not obtain effective return visit content, and the return visit accuracy of the outbound object is improved.
Further, as a specific implementation of fig. 1, an embodiment of the present invention provides a device for revisiting an outbound object, and as shown in fig. 3, the device includes: an acquisition unit 31, a first determination unit 32, an intelligent communication unit 33, a second determination unit 34, and a return access unit 35.
The acquiring unit 31 may be configured to acquire feature data of an outbound object.
The first determining unit 32 may be configured to determine, based on the feature data, a first target call group to which the outbound object belongs.
The intelligent communication unit 33 may be configured to perform intelligent communication with the outbound object based on a first preset terminology corresponding to the first target call group, so as to obtain a returned voice fed back by the outbound object for the first preset terminology.
The second determining unit 34 may be configured to determine, based on the returned speech, a target return visit method corresponding to the outbound object.
The revisit unit 35 may be configured to revisit the outbound object based on the target revisit method.
In a specific application scenario, in order to establish a call group and a preset dialog corresponding to the call group, as shown in fig. 4, the apparatus further includes: a creating unit 36.
The creating unit 36 may be configured to create call groups corresponding to different service scenarios; and creating preset dialogs corresponding to the different call groups.
In a specific application scenario, if the outbound object does not answer the call, the outbound object needs to be called again, based on which the apparatus further includes: a judging unit 37 and a calling unit 38.
The determining unit 37 may be configured to determine whether the call result is an answer call in response to monitoring a call result of the call to the outbound object.
The intelligent communication unit 33 may be specifically configured to, if the call result is an answer call, perform intelligent communication with the outbound object based on a first preset dialect corresponding to the first target call group, to obtain a returned voice fed back by the outbound object for the first preset dialect.
The calling unit 38 may be configured to determine a calling period corresponding to the outbound object if the calling result is that the answering of the call is rejected, and perform a call again on the outbound object based on the calling period.
In a specific application scenario, in order to determine a target return visit method corresponding to the outbound object, the second determining unit 34 includes a word segmentation module 341, a removal module 342, and a first determining module 343.
The word segmentation module 341 may be configured to perform word segmentation processing on the voice text corresponding to the returned voice, so as to obtain each word segmentation corresponding to the voice text.
The removing module 342 may be configured to remove the stop word in each participle, so as to obtain the key participle after the stop word is removed.
The first determining module 343 may be configured to determine a target preset lexicon to which the key participle belongs.
The first determining module 343 may be specifically configured to determine the return visit method corresponding to the target preset lexicon as the target return visit method corresponding to the outbound object.
In a specific application scenario, in order to return the call-out object based on the target return access method, the return access unit 35 includes a forwarding module 351 and a storage module 352.
The forwarding module 351 may be configured to forward the artificial customer service corresponding to the first target call group to communicate with the outbound object if the target preset thesaurus is a first preset thesaurus.
The storage module 352 may be configured to store the outbound object in a blacklist and stop returning the outbound object if the target preset lexicon is a second preset lexicon.
In a specific application scenario, in order to return the call-out object, the return unit 35 further includes a second determining module 353 and a communication module 354.
The second determining module 353 may be configured to, if the target preset thesaurus is a third preset thesaurus, re-determine the second target call group to which the outbound object belongs based on the key participle.
The communication module 354 may be configured to perform intelligent communication with the outbound object based on a second preset dialect corresponding to the second target call group.
In a specific application scenario, in order to determine the first target call group to which the outbound object belongs, the first determining unit 32 may be specifically configured to input the feature data into a preset group division model for group division, so as to obtain the first target call group to which the outbound object belongs.
It should be noted that other corresponding descriptions of the functional modules involved in the device for revisiting an outbound object provided in the embodiment of the present invention may refer to the corresponding description of the method shown in fig. 1, and are not described herein again.
Based on the method shown in fig. 1, correspondingly, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the following steps: acquiring characteristic data of an outbound object; determining a first target call group to which the outbound object belongs based on the feature data; based on a first preset dialect corresponding to the first target calling group, carrying out intelligent communication with the outbound object to obtain a returned voice fed back by the outbound object aiming at the first preset dialect; determining a target return visit method corresponding to the outbound object based on the returned voice; and based on the target return visit method, the return visit is carried out on the outbound object.
Based on the above embodiments of the method shown in fig. 1 and the apparatus shown in fig. 3, an embodiment of the present invention further provides an entity structure diagram of a computer device, as shown in fig. 5, where the computer device includes: a processor 41, a memory 42, and a computer program stored on the memory 42 and executable on the processor, wherein the memory 42 and the processor 41 are each arranged on a bus 43 and the processor 41 implements the following steps when executing the program: acquiring characteristic data of an outbound object; determining a first target call group to which the outbound object belongs based on the feature data; based on a first preset dialect corresponding to the first target call group, performing intelligent communication with the outbound object to obtain returned voice fed back by the outbound object aiming at the first preset dialect; determining a target return visit method corresponding to the outbound object based on the return voice; and based on the target return visit method, returning visits to the outbound object.
According to the technical scheme, the characteristic data of the outbound object is acquired; determining a first target calling group to which the outbound object belongs based on the characteristic data; meanwhile, based on a first preset dialect corresponding to the first target calling group, carrying out intelligent communication with the outbound object to obtain a returned voice fed back by the outbound object aiming at the first preset dialect; then, based on the returned voice, determining a target return visit method corresponding to the outbound object; finally, based on the target return visit method, the outbound object is returned, so that a first target calling group corresponding to the outbound object is determined, communication is carried out with the outbound object based on a first preset conversation corresponding to the first target calling group, then a target return visit method corresponding to the outbound object is determined based on returned voice obtained through the communication, and finally, the outbound object is returned based on the target return visit method, so that the situation that a user is called for many times to cause boredom of the user can be avoided, meanwhile, the time consumed by calling the user for many times can be reduced, meanwhile, the call with the same content can be avoided from being called for many times, the client who wants to handle the service does not obtain effective return visit content, and the return visit accuracy of the outbound object is improved.
It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized in a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a memory device and executed by a computing device, and in some cases, the steps shown or described may be executed out of order, or separately as individual integrated circuit modules, or multiple modules or steps thereof may be implemented as a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made without departing from the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. A method of revisiting an outbound object, comprising:
acquiring characteristic data of an outbound object;
determining a first target call group to which the outbound object belongs based on the feature data;
based on a first preset dialect corresponding to the first target calling group, carrying out intelligent communication with the outbound object to obtain a returned voice fed back by the outbound object aiming at the first preset dialect;
determining a target return visit method corresponding to the outbound object based on the returned voice;
and based on the target return visit method, the return visit is carried out on the outbound object.
2. The method of claim 1, wherein prior to said determining a first target call group to which said outbound subject belongs based on said characterizing data, said method further comprises:
establishing corresponding call groups for different service scenes;
and creating preset dialogs corresponding to the different call groups.
3. The method according to claim 1, wherein before the intelligently communicating with the outbound object based on the first predetermined terminology corresponding to the first target call group to obtain the returned speech fed back by the outbound object for the first predetermined terminology, the method further comprises:
responding to a call result of calling the outbound object, and judging whether the call result is an answering call;
if the calling result is an answering call, intelligently communicating with the outbound object based on a first preset dialect corresponding to the first target calling group to obtain returned voice fed back by the outbound object aiming at the first preset dialect;
and if the calling result is that the answering of the call is refused, determining a calling period corresponding to the outbound object, and calling the outbound object again based on the calling period.
4. The method according to claim 1, wherein the determining a target return visit method corresponding to the outbound object based on the returned speech comprises:
performing word segmentation processing on the voice text corresponding to the returned voice to obtain each word segmentation corresponding to the voice text;
removing stop words in each participle to obtain key participles after the stop words are removed;
determining a target preset word bank to which the key word segmentation belongs;
and determining a return visit method corresponding to the target preset word bank as a target return visit method corresponding to the outbound object.
5. The method of claim 4, wherein the revisiting the outbound object based on the target revisiting method comprises:
if the target preset word bank is a first preset word bank, switching the artificial customer service corresponding to the first target calling group to communicate with the outbound object;
and if the target preset word bank is a second preset word bank, storing the outbound object to a blacklist, and stopping the return visit of the outbound object.
6. The method of claim 4, wherein the revisiting the outbound object based on the target revisiting method comprises:
if the target preset word bank is a third preset word bank, re-determining a second target call group to which the outbound object belongs based on the key participle;
and carrying out intelligent communication with the outbound object based on a second preset conversation corresponding to the second target calling group.
7. The method of claim 1, wherein said determining a first target call group to which the outbound object belongs based on the feature data comprises:
and inputting the characteristic data into a preset group division model for group division to obtain a first target calling group to which the outbound object belongs.
8. A device for revisiting an outbound subject, comprising:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring characteristic data of an outbound object;
a first determining unit, configured to determine, based on the feature data, a first target call group to which the outbound object belongs;
the intelligent communication unit is used for carrying out intelligent communication with the outbound object based on a first preset speech technique corresponding to the first target call group to obtain returned voice fed back by the outbound object aiming at the first preset speech technique;
a second determining unit, configured to determine, based on the returned speech, a target return visit method corresponding to the outbound object;
and the return visit unit is used for returning visits to the outbound object based on the target return visit method.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
10. A computer arrangement comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the computer program realizes the steps of the method of any one of claims 1 to 7 when executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210736358.5A CN115168554A (en) | 2022-06-27 | 2022-06-27 | Callback object return visit method and device, storage medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210736358.5A CN115168554A (en) | 2022-06-27 | 2022-06-27 | Callback object return visit method and device, storage medium and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115168554A true CN115168554A (en) | 2022-10-11 |
Family
ID=83486982
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210736358.5A Pending CN115168554A (en) | 2022-06-27 | 2022-06-27 | Callback object return visit method and device, storage medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115168554A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117714603A (en) * | 2024-02-01 | 2024-03-15 | 济南云上电子科技有限公司 | Outbound method, outbound device and readable storage medium |
| CN117714603B (en) * | 2024-02-01 | 2024-04-30 | 济南云上电子科技有限公司 | Outbound method, outbound device and readable storage medium |
-
2022
- 2022-06-27 CN CN202210736358.5A patent/CN115168554A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117714603A (en) * | 2024-02-01 | 2024-03-15 | 济南云上电子科技有限公司 | Outbound method, outbound device and readable storage medium |
| CN117714603B (en) * | 2024-02-01 | 2024-04-30 | 济南云上电子科技有限公司 | Outbound method, outbound device and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10262195B2 (en) | Predictive and responsive video analytics system and methods | |
| WO2020258654A1 (en) | Answer acquisition method and device | |
| US9930175B2 (en) | Systems and methods for lead routing | |
| CN110138982A (en) | Service based on artificial intelligence is realized | |
| CN111026843A (en) | Artificial intelligent voice outbound method, system and storage medium | |
| CN108073600A (en) | A kind of intelligent answer exchange method, device and electronic equipment | |
| WO2014145149A1 (en) | Intelligent automated agent for a contact center | |
| CN110457709A (en) | Outgoing call dialog process method, apparatus and server | |
| CN110942326A (en) | User satisfaction prediction method and device and server | |
| CN106936971B (en) | A kind of incoming person's information presentation system and reminding method | |
| CN110110038A (en) | Traffic predicting method, device, server and storage medium | |
| CN111597818A (en) | Call quality inspection method, call quality inspection device, computer equipment and computer readable storage medium | |
| CN107800901A (en) | User's call processing method, device, computer equipment and storage medium | |
| CN113724036A (en) | Method and electronic equipment for providing question consultation service | |
| CN113159901A (en) | Method and device for realizing financing lease service session | |
| CN115168554A (en) | Callback object return visit method and device, storage medium and computer equipment | |
| CN115083412B (en) | Voice interaction method and related device, electronic equipment and storage medium | |
| CN113873087B (en) | Outbound method, outbound device, computer equipment and storage medium | |
| Hepenstal et al. | How analysts think: A preliminary study of human needs and demands for AI-based conversational agents | |
| CN109862195A (en) | Intelligent sound calling system and method based on information automatic error-correcting | |
| KR20160104462A (en) | Device and System for providing phone number service by providing advertisements using emotion analysis of customer and method thereof | |
| US11196864B1 (en) | Analyzing voice response to telephone call to assign appropriate agent | |
| CN113905135A (en) | User intention identification method and device of intelligent outbound robot | |
| CN112333341A (en) | Intelligent voice robot system | |
| CN113327582B (en) | Voice interaction method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |